Showing papers by "Byron M. Yu published in 2011"
Proceedings Article•
12 Dec 2011
TL;DR: This work argues that in the cortex, where firing exhibits extensive correlations in both time and space and where a typical sample of neurons still reflects only a very small fraction of the local population, the most appropriate model captures shared variability by a low-dimensional latent process evolving with smooth dynamics, rather than by putative direct coupling.
Abstract: Neurons in the neocortex code and compute as part of a locally interconnected population. Large-scale multi-electrode recording makes it possible to access these population processes empirically by fitting statistical models to unaveraged data. What statistical structure best describes the concurrent spiking of cells within a local network? We argue that in the cortex, where firing exhibits extensive correlations in both time and space and where a typical sample of neurons still reflects only a very small fraction of the local population, the most appropriate model captures shared variability by a low-dimensional latent process evolving with smooth dynamics, rather than by putative direct coupling. We test this claim by comparing a latent dynamical model with realistic spiking observations to coupled generalised linear spike-response models (GLMs) using cortical recordings. We find that the latent dynamical approach outperforms the GLM in terms of goodness-of-fit, and reproduces the temporal correlations in the data more accurately. We also compare models whose observations models are either derived from a Gaussian or point-process models, finding that the non-Gaussian model provides slightly better goodness-of-fit and more realistic population spike counts.
233 citations
TL;DR: The initial condition hypothesis elucidates a view of the relationship between single-trial preparatory neural population dynamics and single- trial behavior and is shown to explain approximately 4-fold more arm-movement reaction-time variance than the best alternative method.
217 citations
Proceedings Article•
12 Dec 2011
TL;DR: A Hidden Switching Linear Dynamical Systems model is shown to be able to distinguish different dynamical regimes within single-trial motor cortical activity associated with the preparation and initiation of hand movements and performs better than recent comparable models in predicting the firing rate of an isolated neuron based on the firing rates of others.
Abstract: Simultaneous recordings of many neurons embedded within a recurrently-connected cortical network may provide concurrent views into the dynamical processes of that network, and thus its computational function. In principle, these dynamics might be identified by purely unsupervised, statistical means. Here, we show that a Hidden Switching Linear Dynamical Systems (HSLDS) model— in which multiple linear dynamical laws approximate a nonlinear and potentially non-stationary dynamical process—is able to distinguish different dynamical regimes within single-trial motor cortical activity associated with the preparation and initiation of hand movements. The regimes are identified without reference to behavioural or experimental epochs, but nonetheless transitions between them correlate strongly with external events whose timing may vary from trial to trial. The HSLDS model also performs better than recent comparable models in predicting the firing rate of an isolated neuron based on the firing rates of others, suggesting that it captures more of the "shared variance" of the data. Thus, the method is able to trace the dynamical processes underlying the coordinated evolution of network activity in a way that appears to reflect its computational role.
73 citations
Patent•
17 Feb 2011TL;DR: In this article, a modified brain machine interface is developed by modifying the vectors of the velocities defined in the brain machine interfaces, which can now be used to control a prosthetic device using recorded neural brain signals from a user of the prosthetic devices.
Abstract: Artificial control of a prosthetic device is provided. A brain machine interface contains a mapping of neural signals and corresponding intention estimating kinematics (e.g. positions and velocities) of a limb trajectory. The prosthetic device is controlled by the brain machine interface. During the control of the prosthetic device, a modified brain machine interface is developed by modifying the vectors of the velocities defined in the brain machine interface. The modified brain machine interface includes a new mapping of the neural signals and the intention estimating kinematics that can now be used to control the prosthetic device using recorded neural brain signals from a user of the prosthetic device. In one example, the intention estimating kinematics of the original and modified brain machine interface includes a Kalman filter modeling velocities as intentions and positions as feedback.
7 citations
23 Jun 2011
TL;DR: The utility of instructed paths for pushing the limits of the subject's control and rigorously quantifying the accuracy of cursor movements are demonstrated, both of which are critical for increasing the clinical viability of neural prosthetic systems.
Abstract: Neural prostheses are becoming increasingly feasible as assistive technologies for paralyzed patients. A major goal is to provide control of a prosthesis rivaling the natural arm in speed, accuracy, and flexibility. Here, we demonstrate high-performance cursor control by training a monkey to move a cursor in a 2D virtual reality environment using neural activity recorded in primary motor cortex. On a standard center-out task with 8 possible targets, the subject maintained a success rate greater than 95% over many hundreds of trials, on par with previous reports. We introduced the more challenging task of moving the cursor along instructed paths with zero, one, and two inflections. Over several weeks, the subject's performance with double-inflection paths reached a stable level of greater than 55% success with movement times approaching those of the natural arm. Our instructed trajectory task provides a new standard for quantification of prosthesis performance: since the subject's intended movement is known (i.e. the instructed path), we can compute the root mean-square-error (RMSE) between the decoded and intended cursor position throughout the reach. We found that, while success rate tended to increase with training, the RMSE among successful trials remained largely unchanged, consistent with the all-or-none reward scheme. In sum, this work demonstrates the utility of instructed paths for i) pushing the limits of the subject's control and ii) rigorously quantifying the accuracy of cursor movements, both of which are critical for increasing the clinical viability of neural prosthetic systems.
5 citations
01 Jan 2011
TL;DR: Scaling of a Newly Learned Sensorimotor Transformation Reorganization of Finger Coordination Patterns During Adaptation to Rotation and Hypergeneralization of Learned Dynamics Across Movement Speeds Reveals.
Abstract: [PDF] [Full Text] [Abstract] , April , 2010; 103 (4): 2124-2138. J Neurophysiol Joo-Hyun Song and Robert M. McPeek Selection and Movement Production Roles of Narrowand Broad-Spiking Dorsal Premotor Area Neurons in Reach Target [PDF] [Full Text] [Abstract] , May 18, 2010; 10 (5): . J Vis Fabrice R. Sarlegna and Jean Blouin by amplitude control Visual guidance of arm reaching: Online adjustments of movement direction are impaired [PDF] [Full Text] [Abstract] , August , 2010; 104 (2): 799-810. J Neurophysiol Stephen I. Ryu and Krishna V. Shenoy Matthew T. Kaufman, Mark M. Churchland, Gopal Santhanam, Byron M. Yu, Afsheen Afshar, Roles of Monkey Premotor Neuron Classes in Movement Preparation and Execution [PDF] [Full Text] [Abstract] , January , 2011; 105 (1): 454-473. J Neurophysiol Scheidt Xiaolin Liu, Kristine M. Mosier, Ferdinando A. Mussa-Ivaldi, Maura Casadio and Robert A. Scaling of a Newly Learned Sensorimotor Transformation Reorganization of Finger Coordination Patterns During Adaptation to Rotation and [PDF] [Full Text] [Abstract] , January , 2011; 105 (1): 45-59. J Neurophysiol Wilsaan M. Joiner, Obafunso Ajayi, Gary C. Sing and Maurice A. Smith Anisotropic, Gain-Encoding Primitives for Motor Adaptation Linear Hypergeneralization of Learned Dynamics Across Movement Speeds Reveals